Skip to content

pixano_inference.ray.routes.service

Service endpoints: health, readiness, settings, model listing.

register_service_routes(app, deployment_manager)

Register health, readiness, settings, and model listing endpoints.

Parameters:

Name Type Description Default
app FastAPI

FastAPI application.

 required
deployment_manager DeploymentManager

The deployment manager instance.

 required
Source code in pixano_inference/ray/routes/service.py 
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
def register_service_routes(app: FastAPI, deployment_manager: DeploymentManager) -> None:
    """Register health, readiness, settings, and model listing endpoints.

    Args:
        app: FastAPI application.
        deployment_manager: The deployment manager instance.
    """

    @app.get("/health")
    async def health():
        """Health check endpoint."""
        return {"status": "healthy"}

    @app.get("/ready")
    async def ready():
        """Readiness check endpoint."""
        model_count = len(deployment_manager.list_models())
        return {
            "ready": True,
            "models_loaded": model_count,
            "version": __version__,
        }

    @app.get("/")
    async def root():
        """Root endpoint."""
        return {
            "message": "Pixano Inference API (Ray Serve)",
            "version": __version__,
            "docs": "/docs",
        }

    @app.get("/app/settings/")
    async def get_settings():
        """Get application settings and status."""
        gpu_info = deployment_manager.get_gpu_info()
        models = deployment_manager.list_models()
        used_gpus = gpu_info.get("gpus_used", 0)
        num_gpus = gpu_info.get("num_gpus", 0)
        return {
            "app_name": "Pixano Inference",
            "app_version": __version__,
            "app_description": "Pixano Inference API powered by Ray Serve",
            "num_cpus": deployment_manager.config.num_cpus,
            "num_gpus": num_gpus,
            "num_nodes": 1,
            "gpus_used": used_gpus,
            "gpu_to_model": {},
            "models": [m.name for m in models],
            "models_to_capability": {m.name: m.capability for m in models},
        }

    @app.get("/app/models/")
    async def list_models() -> list[ModelInfo]:
        """List all deployed models."""
        return deployment_manager.list_models()